Gene prediction method, electronic device and application thereof for improving gene UTR region prediction completeness

Abstract

The present invention provides a method, an electronic device and an application thereof for improving the prediction integrity of a gene UTR region. The method comprises: S1) performing a first transcriptome prediction, a second transcriptome prediction, a first homology prediction and a third transcriptome prediction on the genome of a sample to obtain the corresponding first initial prediction gene set, a second initial prediction gene set, a third initial prediction gene set and a fourth initial prediction gene set respectively; S2) integrating the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set. The method can solve the problem of low gene prediction integrity in the UTR region of the prior art and is applicable to the field of gene prediction.

Description

Technical Field

[0001] The present invention relates to the field of gene prediction, and in particular to a gene prediction method, an electronic device and an application thereof for improving the prediction integrity of a gene UTR region. Background Art

[0002] The accuracy of genome annotation is crucial for biological research. It is not only related to the understanding of gene function, but also directly affects the analysis of disease mechanisms, the discovery of drug targets and the application of biotechnology. An inaccurate genome annotation result may lead to wrong biological conclusions, waste valuable research resources and may mislead future research directions.

[0003] The UTR (Untranslated Region), especially the 3'UTR and 5'UTR, is not only involved in the regulation of gene expression, but also involves key biological processes such as mRNA stability, translation efficiency, and protein degradation. Mutations in these regions are associated with a variety of diseases, including cancer, neurodegenerative diseases, and cardiovascular diseases. Therefore, in-depth research on the UTR region is of great significance for revealing the molecular mechanisms of the disease and developing new treatment strategies. However, at present, conventional gene prediction methods using existing technologies can usually only predict the UTR regions of a small number of genes, resulting in low integrity of the UTR regions in the predicted gene set, which has hindered gene research involving the UTR region.

[0004] Therefore, developing a new gene prediction method that can improve the completeness of gene UTR region prediction is of great significance for improving the overall integrity of genome annotation, deepening the understanding of gene regulatory mechanisms, and promoting the progress of biomedical research. Summary of the invention

[0005] The main purpose of the present invention is to provide a gene prediction method, an electronic device and its application for improving the prediction integrity of the UTR region of the gene, so as to solve the problem of low gene prediction integrity in the UTR region of the prior art.

[0006] In order to achieve the above-mentioned purpose, according to a first aspect of the present invention, a method for gene prediction for improving the completeness of gene UTR region prediction is provided, the method comprising: S1) performing a first transcriptome prediction, a second transcriptome prediction, a first homology prediction and a third transcriptome prediction on the genome of the sample, respectively, to obtain a corresponding first initial prediction gene set, a second initial prediction gene set, a third initial prediction gene set and a fourth initial prediction gene set, respectively; wherein the first initial prediction gene set is obtained by performing a first transcriptome prediction on the genome of the sample using the second-generation transcript data of the sample; the second initial prediction gene set is obtained by performing a second transcriptome prediction on the genome of the sample using the third-generation transcript data of the sample; S2) integrating the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set.

[0007] Furthermore, the third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using a gene set in a homologous species genome database; preferably, the software for the first homology prediction includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

[0008] Furthermore, the fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using a full-length transcript reference genome; preferably, the full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

[0009] Furthermore, the software for performing the first transcriptome prediction, the software for performing the second transcriptome prediction, and the software for performing the third transcriptome prediction are each independently selected from PASA software, and / or transdecoder software.

[0010] Further, S2) includes: S21) comparing the first initial predicted gene set with the initial gene set to be compared to obtain a first overlapping region gene set; performing a first integration on the first overlapping region gene set to obtain a first predicted gene set; wherein the method for obtaining the initial gene set to be compared is: performing sample gene prediction on the genome of the sample to obtain an initial gene set to be compared; S22) comparing the first predicted gene set with the second initial predicted gene set to obtain a second overlapping region gene set; performing a second integration on the second overlapping region gene set to obtain a second predicted gene set; S23) comparing the second predicted gene set with the third initial predicted gene set to obtain a third overlapping region gene set; performing a third integration on the third overlapping region gene set to obtain a third predicted gene set; S24) comparing the third predicted gene set with the fourth initial predicted gene set to obtain a fourth overlapping region gene set; performing a fourth integration on the fourth overlapping region gene set to obtain a final predicted gene set.

[0011] Further, the initial gene set to be compared, the second initial predicted gene set, the third initial predicted gene set and the fourth initial predicted gene set are recorded as the gene sets to be compared; the first initial predicted gene set, the first predicted gene set, the second predicted gene set and the third predicted gene set are recorded as the reference gene sets; the first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set are recorded as overlapping region gene sets;

[0012] Integration includes:

[0013] a) Compare the gene set to be compared with the reference gene set to obtain the overlapping region gene set; filter the gene pairs in the overlapping region gene set that have inconsistent directions between the gene set to be compared and the reference gene set, as well as the genes with UTR regions at both ends in the reference gene set, in the gene set to be compared to obtain the filtered gene set;

[0014] b) Extend each gene in the filtered gene set to obtain the extended region of each gene; the extended region does not overlap with the position of the gene in the upstream region or the position of the gene in the downstream region of the gene corresponding to the reference gene set;

[0015] Wherein, one of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region, which is extended to the other end of the gene in the filtered gene set to obtain an extended region;

[0016] The genes in the filtered gene set do not contain UTR regions at both ends. The genes in the reference gene set corresponding to the genes in the filtered gene set have UTR regions at both ends. The extended regions are extended to both ends of the genes:

[0017] The extension includes: when the starting position of a gene in the reference gene set is smaller than the starting position of a corresponding gene in the filtering gene set, extending upstream of the gene in the filtering gene set to obtain an extended region;

[0018] When the end position of a gene in the reference gene set is smaller than the start position of a gene in the corresponding filtering gene set, and the end position of a gene in the reference gene set is smaller than the end position of a gene in the corresponding filtering gene set, the gene is extended downstream of the gene in the filtering gene set to obtain an extended region;

[0019] c) Adding the information of genes in the extended region to the reference gene set to obtain the final predicted gene set; the information of genes in the extended region includes the sequence position and size of the genes.

[0020] According to a second aspect of the present invention, there is provided an electronic device for gene prediction for improving the completeness of gene UTR region prediction, the electronic device comprising an initial prediction unit and an integration unit;

[0021] The initial prediction unit is used to perform a first transcriptome prediction, a second transcriptome prediction, a first homology prediction and a third transcriptome prediction on the genome of the sample, respectively, to obtain a corresponding first initial prediction gene set, a second initial prediction gene set, a third initial prediction gene set and a fourth initial prediction gene set;

[0022] The first initial prediction gene set is obtained by performing a first transcriptome prediction on the sample genome using the second-generation transcript data of the sample; the second initial prediction gene set is obtained by performing a second transcriptome prediction on the sample genome using the third-generation transcript data of the sample;

[0023] The integration unit is used to integrate the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set.

[0024] Furthermore, the third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using a gene set in a homologous species genome database; preferably, the software for the first homology prediction includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

[0025] Further, the fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using the full-length transcript reference genome;

[0026] Preferably, the full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

[0027] Furthermore, the software for performing the first transcriptome prediction, the software for performing the second transcriptome prediction, and the software for performing the third transcriptome prediction are each independently selected from PASA software and / or transdecoder software.

[0028] Further, the integration unit includes a first prediction gene set acquisition unit, a second prediction gene set acquisition unit, a third prediction gene set acquisition unit, and a prediction gene set acquisition unit;

[0029] The first prediction gene set acquisition unit is used to compare the first initial prediction gene set with the initial gene set to be compared in the database to obtain a first overlapping region gene set; perform a first integration on the first overlapping region gene set to obtain a first prediction gene set;

[0030] The method for obtaining the initial gene set to be compared is as follows: performing sample gene prediction on the genome of the sample to obtain the initial gene set to be compared;

[0031] A second prediction gene set acquisition unit is used to compare the first prediction gene set with the second initial prediction gene set to obtain a second overlapping region gene set; perform a second integration on the second overlapping region gene set to obtain a second prediction gene set;

[0032] A third prediction gene set acquisition unit is used to compare the second prediction gene set with the third initial prediction gene set to obtain a third overlapping region gene set; perform a third integration on the third overlapping region gene set to obtain a third prediction gene set;

[0033] The prediction gene set acquisition unit is configured to compare the third prediction gene set with the fourth initial prediction gene set to obtain a fourth overlapping region gene set; and perform a fourth integration on the fourth overlapping region gene set to obtain a final prediction gene set.

[0034] Furthermore, the electronic device also includes: a first labeling unit, used to record the initial comparison gene, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set as the gene sets to be compared; a second labeling unit, used to record the first initial prediction gene set, the first prediction gene set, the second prediction gene set and the third prediction gene set as the reference gene set; a third labeling unit, used to record the first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set as the overlapping region gene set;

[0035] The integration unit includes: a comparison module, which is used to compare the gene set to be compared with the reference gene set to obtain the gene set of the overlapping region;

[0036] A filtering module is used to filter the gene pairs with inconsistent directions between the to-be-compared gene set and the reference gene set in the overlapping region gene set, as well as the genes with UTR regions at both ends in the reference gene set, in the to-be-compared gene set to obtain a filtered gene set;

[0037] An extension module is used to extend each gene in the filtered gene set to obtain the extended region of each gene;

[0038] The extended region does not overlap with the positions of genes in the upstream region or the downstream region of the genes corresponding to the reference gene set;

[0039] The extended modules include:

[0040] The first extension submodule is used for extending the gene in the filtered gene set to the other end of the gene in the filtered gene set to obtain an extended region when one of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region;

[0041] The second extension submodule is used to extend to both ends of the gene to obtain the extended region when both ends of the gene in the filtered gene set do not contain UTR regions and both ends of the gene in the reference gene set corresponding to the gene in the filtered gene set have UTR regions:

[0042] A third extension submodule is used to extend upstream of the gene in the filtering gene set to obtain an extended region when the starting position of the gene in the reference gene set is smaller than the starting position of the corresponding gene in the filtering gene set;

[0043] A fourth extension submodule, for extending downstream of the gene in the filtering gene set to obtain an extension region when the end position of the gene in the reference gene set is smaller than the start position of the gene in the corresponding filtering gene set and the end position of the gene in the reference gene set is smaller than the end position of the gene in the corresponding filtering gene set;

[0044] Integrator module: Add the information of genes in the extended region to the reference gene set to obtain the final predicted gene set; the information of genes in the extended region includes the sequence position and size of the genes.

[0045] In a third typical embodiment of the present application, a computer-readable storage medium is provided, which includes a stored program, wherein when the program is executed, the above-mentioned gene prediction method for improving the completeness of gene UTR region prediction is controlled.

[0046] In a fourth typical embodiment of the present application, a processor is provided, the processor being used to run a program, wherein the program executes the above-mentioned gene prediction method for improving the completeness of gene UTR region prediction when running.

[0047] Apply the technical scheme of the present invention, the genome of the sample is sequentially subjected to the first transcriptome prediction, the second transcriptome prediction, the first homology prediction and the third transcriptome prediction, and the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set are obtained, wherein the first transcriptome prediction and the second transcriptome prediction are based on the gene prediction of the genome of the sample based on the second-generation transcript sequencing data and the third-generation transcript sequencing data, respectively. Then, the above-mentioned different initial prediction gene sets are integrated to obtain the predicted gene set finally used for gene prediction, which contains the gene set obtained by using the existing gene prediction means, combines the advantages of the second-generation sequencing and the third-generation sequencing, and predicts that the number of UTR regional structure annotations in the gene set increases, further improving the accuracy of the UTR prediction results, and improving the overall accuracy of gene prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0049] Figure 1 A schematic flow chart of a method for gene prediction for improving the completeness of gene UTR region prediction according to an embodiment of the present invention is shown.

[0050] Figure 2 A schematic diagram of an electronic device for a gene prediction method for improving the completeness of gene UTR region prediction according to an embodiment of the present invention is shown.

[0051] Figure 3 A schematic diagram of a method for obtaining an extended region of gene prediction for improving the completeness of gene UTR region prediction according to Example 1 of the present invention is shown.

[0052] Figure 4 A hardware structure block diagram of a gene prediction method for improving the completeness of gene UTR region prediction according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0053] It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present invention will be described in detail below in conjunction with the embodiments. As mentioned in the background technology, only a small amount of genes in the UTR region can be predicted in the prior art, so that the gene prediction integrity in the UTR region is relatively low. Therefore, in the present application, the inventor attempts to develop a new method for gene prediction for improving the gene UTR region prediction integrity, which can predict a large number of UTR region genes, and can further improve the overall comprehensiveness and integrity of gene prediction, thereby proposing a series of protection schemes of the present application.

[0054] In a first typical embodiment of the present application, a method for gene prediction for improving the completeness of gene UTR region prediction is provided (the method flow diagram is as follows: Figure 1 As shown), the method comprises:

[0055] S1) The genome of the sample is respectively subjected to the first transcriptome prediction, the second transcriptome prediction, the first homology prediction and the third transcriptome prediction to obtain the corresponding first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set; wherein, the first initial prediction gene set is obtained by performing the first transcriptome prediction on the genome of the sample using the second-generation transcript data of the sample; the second initial prediction gene set is obtained by performing the second transcriptome prediction on the genome of the sample using the third-generation transcript data of the sample; S2) The first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set are integrated to obtain the final prediction gene set.

[0056] In the prior art, the gene prediction methods of the prior art, including the gene prediction methods based on the second-generation transcriptome and the gene prediction methods based on the third-generation transcriptome, can usually only predict the UTR regions of a small number of genes, resulting in low gene integrity in the UTR region. Moreover, the prior art usually uses the function in the PASA software program (https: / / github.com / PASApipeline / PASApipeline / wiki / PASA_genome_annotation) to add UTR region gene prediction results, but the number of UTRs that can be added by this function is also very limited. Therefore, the prior art has low integrity in predicting genes in the UTR region, which in turn affects subsequent gene expression analysis, protein interaction network construction, and analysis of signal transduction pathways.

[0057] Generally speaking, the read length of sample sequencing data obtained by the second-generation transcript sequencing technology is relatively short, but high-throughput sequencing can provide a large amount of sample transcript data, covering various regions of the transcriptome, and providing rich information for the prediction of UTR regions. In large-scale gene expression analysis, the second-generation sequencing technology can quickly and economically obtain gene expression profiles, and has broad prospects for applications such as disease diagnosis and drug screening. However, the transcript sequences obtained by the second-generation sequencing are relatively short, which can easily lead to splicing errors, resulting in reduced integrity of the UTR region genes in the gene prediction results, resulting in reduced accuracy of overall gene prediction.

[0058] The sequencing of the third-generation transcript sequencing technology can obtain sample transcript data with ultra-long reads. This ultra-long read can cover the entire transcript of the sample, including the UTR region, avoiding sequence splicing errors caused by too short reads, and can further improve the completeness of UTR region predictions. In the annotation of complex regions of the genome, the third-generation sequencing technology can help solve the problems of repetitive sequences and complex structures that are difficult to handle with short-read sequencing technology. However, the ultra-long read data obtained by the third-generation transcript sequencing has the possibility of mosaicism, resulting in too many repetitive sequences. Therefore, if only the third-generation transcript data is used for gene prediction, there is also the possibility of reducing the completeness of gene prediction in the UTR region.

[0059] This application combines the advantages and characteristics of second-generation sequencing and third-generation sequencing, and uses second-generation transcript data and third-generation transcript data to predict the genome of the sample, combining homology prediction and transcriptome prediction methods to obtain initial prediction gene sets from multiple sources, and optimize and integrate these initial prediction gene sets to obtain gene prediction sets for multiple UTR regions of the sample. Compared with the prior art, the gene prediction method of the present application improves the prediction integrity of the UTR region of the sample, and further improves the overall accuracy of gene prediction.

[0060] Before the second-generation transcript sequencing data of the sample is used for the first transcriptome prediction, the second-generation transcript sequencing data of the sample is compared using comparison software including but not limited to hisat2, STAR or Bowtie2 to ensure the accuracy of subsequent data processing. Then, the sequencing numbers after the comparison are assembled using assembly software including but not limited to Stringtie, Trinity or Cufflinks to obtain complete second-generation transcript data. The complete second-generation transcript data and the genome of the above-mentioned sample are used for the first transcriptome prediction to obtain the above-mentioned first initial prediction gene set. Before the third-generation transcript sequencing data of the sample is used for the second transcriptome prediction, the third-generation transcript data of the sample is first clustered and fully processed to remove chimeric fragments to ensure that repeated fragments appear during subsequent annotations, increasing the burden of data processing.

[0061] In a preferred embodiment, the third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using the genome of the homologous species genome sample; preferably, the software for the first homology prediction includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

[0062] This application maps the annotation gene set of homologous species and the reference genome to the genome of the sample to further annotate its UTR region. Those skilled in the art can flexibly select any homology prediction means or the genome of the homologous species sample according to the actual situation, and the homology prediction of the genome of the sample can achieve the effect of this application. The gene annotation information of homologous species and the full-length transcript sequence in the existing database provide additional information sources for prediction, which can make up for the deficiency of sample data and improve the reliability of prediction results.

[0063] In a preferred embodiment, the fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using a full-length transcript reference genome; preferably, the full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

[0064] This application uses the gene set of high-quality transcripts collected by the database in the prior art to perform a third transcriptome prediction on the gene set of the sample, and further predicts and annotates the genes in the UTR region of the sample; it is conducive to improving the overall prediction completeness and accuracy. Those skilled in the art can flexibly select any existing transcriptome prediction method or gene set in the transcript database according to actual conditions, and the third transcriptome prediction of the genome of the sample can achieve the effect of this application.

[0065] Each step of the above gene prediction method can be optimized and adjusted in detail according to the material of the sample and the research needs. For example, the alignment software of the second-generation sequencing technology can be selected according to the complexity and sequence characteristics of the species genome, and the selection of the read length and splicing algorithm of the third-generation sequencing technology also needs to take into account the biological characteristics of the species and the quality of the sequencing data. In addition, the selection of homologous species and the use of Refseq libraries also need to be considered based on the relationship of the target species and the availability of transcript sequences. Through the flexible application and parameter optimization of this method, the accuracy and completeness of gene UTR region prediction can be significantly improved, providing valuable gene information resources for biological research and applications.

[0066] In a preferred embodiment, S2) includes: S21) comparing the first initial predicted gene set with the initial gene set to be compared to obtain a first overlapping region gene set; performing a first integration on the first overlapping region gene set to obtain a first predicted gene set; wherein the method for obtaining the initial gene set to be compared is: performing sample gene prediction on the genome of the sample to obtain the initial gene set to be compared; S22) comparing the first predicted gene set with the second initial predicted gene set to obtain a second overlapping region gene set; performing a second integration on the second overlapping region gene set to obtain a second predicted gene set; S23) comparing the second predicted gene set with the third initial predicted gene set to obtain a third overlapping region gene set; performing a third integration on the third overlapping region gene set to obtain a third predicted gene set; S24) comparing the third predicted gene set with the fourth initial predicted gene set to obtain a fourth overlapping region gene set; performing a fourth integration on the fourth overlapping region gene set to obtain a final predicted gene set.

[0067] The method for sample gene prediction is the conventional gene prediction method in the prior art, including performing sample transcriptome prediction, sample homology prediction and sample de novo prediction on the sample genome, and obtaining a first sample prediction gene set, a second sample prediction gene set and a third sample prediction gene set respectively; using EVM software to integrate the first sample prediction gene set, the second sample prediction gene set and the third sample prediction gene set, the initial gene set to be compared can be obtained.

[0068] Homology prediction of sample genomes: The proteins of some genes are highly conserved among similar species, so the existing high-quality annotation information of closely related species is used to determine the exon boundaries and splicing sites through sequence comparison during prediction; transcriptome prediction of sample genomes: With the assistance of RNA-seq data of species for annotation, the exon regions and splicing sites can be determined more accurately; de novo prediction of sample genomes: Gene structure is predicted through existing probability models.

[0069] In a preferred embodiment, the initial gene set to be compared, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set are recorded as the gene sets to be compared; the first initial prediction gene set, the first prediction gene set, the second prediction gene set and the third prediction gene set are recorded as the reference gene sets;

[0070] The first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set are recorded as overlapping region gene sets;

[0071] The integration includes: a) comparing the gene set to be compared with the reference gene set to obtain the overlapping region gene set; filtering the gene pairs in the overlapping region gene set that have inconsistent directions between the gene set to be compared and the reference gene set, as well as the genes with UTR regions at both ends of the reference gene set, in the gene set to be compared to obtain the filtered gene set;

[0072] b) Extend each gene in the filtered gene set to obtain the extended region of each gene; the extended region does not overlap with the position of the gene in the upstream region or the position of the gene in the downstream region of the gene corresponding to the reference gene set;

[0073] in,

[0074] One of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region, and the region is extended to the other end of the gene in the filtered gene set to obtain an extended region;

[0075] The genes in the filtered gene set do not contain UTR regions at both ends. The genes in the reference gene set corresponding to the genes in the filtered gene set have UTR regions at both ends. The extended regions are extended to both ends of the genes:

[0076] The extension includes: when the starting position of a gene in the reference gene set is smaller than the starting position of a corresponding gene in the filtering gene set, extending upstream of the gene in the filtering gene set to obtain an extended region;

[0077] When the end position of a gene in the reference gene set is smaller than the start position of a gene in the corresponding filtering gene set, and the end position of a gene in the reference gene set is smaller than the end position of a gene in the corresponding filtering gene set, the gene is extended downstream of the gene in the filtering gene set to obtain an extended region;

[0078] c) Adding the information of genes in the extended region to the reference gene set to obtain the final predicted gene set; the information of genes in the extended region includes the sequence position and size of the genes.

[0079] The schematic diagram of the extended area acquisition method is shown in Figure 3 As shown, “cds1”, “cds2” and “cds3” refer to the three coding regions (CDS, Coding Sequence) of the gene.

[0080] The overlapping region gene set mentioned above refers to the gene set of the overlapped part in two gene sets (reference gene set and gene set to be compared), that is, the set of a gene sequence shared by two or more genes. By judging the UTR situation at both ends of the genes in the reference gene set involved in these overlaps, we first filter out the gene pairs with inconsistent directions in the overlapping region gene set (one of the gene pairs comes from the reference gene set and the other comes from the gene set to be compared), as well as the genes with UTR regions at both ends of the reference gene set, to reduce the burden of subsequent screening of UTR region prediction gene sets.

[0081] Then, the single-end UTR of the gene from the reference gene set in the above gene pair is judged, and then the gene in the filtered gene set is extended to obtain the UTR region.

[0082] When the 5' end of the gene in the reference gene set corresponding to the gene pair in the filtered gene set has a UTR region, the extension is performed toward the 3' end of the gene pair in the filtered gene set to obtain an extended region; when the 3' end of the gene in the reference gene set corresponding to the gene pair in the filtered gene set has a UTR region, the extension is performed toward the 5' end of the gene pair in the filtered gene set to obtain an extended region;

[0083] If the reference gene set has no UTR regions at both ends, the start and end positions of the genes in the overlap region and the start and end positions of the genes in the reference gene set are used to determine and obtain the extended region to obtain the UTR region of the gene. If the reference gene set corresponding to a gene in the filtered gene set has no UTR region, the gene will not be extended.

[0084] The gene structure of the extended region obtained after the above extension not only includes the UTR region but also may include the gene information of exons, coding regions or introns. These genes are merged into the above reference gene set for summary, and then the next round of comparison and integration is carried out to make the final predicted gene set more complete.

[0085] The tools commonly used in the prior art to add UTR region gene prediction results are mainly functions in the PASA software program (https: / / github.com / PASApipeline / PASApipeline / wiki / PASA_genome_annotation), which uses the second-generation transcriptome data of the samples obtained by PASA splicing to compare and update the predicted gene set that has been predicted, integrate the evidence in the PASA alignment results, correct the exon boundaries and add the UTR region, that is, the above-mentioned initial gene set to be compared, but the number of UTRs that can be added by this function is limited. The gene prediction method of the present application combines the second-generation and third-generation sequencing technologies, and uses the information of the full-length transcripts of homologous species and samples in the database to predict genes in the genome of the sample. Through the integration and optimization of multi-source data, the accuracy and completeness of the prediction of the UTR region of the sample gene are significantly improved. It is helpful to more accurately locate the non-coding region of the gene in subsequent studies, which is of great value for understanding gene regulation mechanisms, disease-related gene research, and genetic engineering applications. Moreover, the method of the present application can also be applied to the updating and optimization of genome annotations, which is of great significance for the genome annotations of new species and the improvement of genome annotations of known species, helps to promote the frontier development of genomics and transcriptomics research, and provides data support for research in fields such as biomedicine, agricultural biotechnology and ecology.

[0086] In a second typical embodiment of the present application, an electronic device for gene prediction for improving the completeness of gene UTR region prediction is provided ( Figure 2 is a schematic diagram of the electronic device of the present application), the electronic device comprises an initial prediction unit 10 and an integration unit 20;

[0087] Among them, the initial prediction unit 10 is used to perform a first transcriptome prediction, a second transcriptome prediction, a first homology prediction and a third transcriptome prediction on the genome of the sample, respectively, to obtain the corresponding first initial prediction gene set, second initial prediction gene set, third initial prediction gene set and fourth initial prediction gene set respectively; the first initial prediction gene set is obtained by performing a first transcriptome prediction on the genome of the sample using the second-generation transcript data of the sample; the second initial prediction gene set is obtained by performing a second transcriptome prediction on the genome of the sample using the third-generation transcript data of the sample; the integration unit 20 is used to integrate the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain the final prediction gene set;

[0088] The integration unit 20 is used to integrate the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set.

[0089] In a preferred embodiment, the third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using the genome of the homologous species genome sample; preferably, the software for the first homology prediction includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

[0090] In a preferred embodiment, the fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using a full-length transcript reference genome; preferably, the full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

[0091] In a preferred embodiment, the software for performing the first transcriptome prediction, the software for performing the second transcriptome prediction, and the software for performing the third transcriptome prediction are each independently selected from PASA software and / or transdecoder software.

[0092] In a preferred embodiment, the integration unit includes a first prediction gene set acquisition unit, a second prediction gene set acquisition unit, a third prediction gene set acquisition unit, and a prediction gene set acquisition unit;

[0093] The first prediction gene set acquisition unit is used to compare the first initial prediction gene set with the initial gene set to be compared in the database to obtain a first overlapping region gene set; perform a first integration on the first overlapping region gene set to obtain a first prediction gene set;

[0094] The method for obtaining the initial gene set to be compared is as follows: performing sample gene prediction on the genome of the sample to obtain the initial gene set to be compared;

[0095] A second prediction gene set acquisition unit is used to compare the first prediction gene set with the second initial prediction gene set to obtain a second overlapping region gene set; perform a second integration on the second overlapping region gene set to obtain a second prediction gene set;

[0096] A third prediction gene set acquisition unit is used to compare the second prediction gene set with the third initial prediction gene set to obtain a third overlapping region gene set; perform a third integration on the third overlapping region gene set to obtain a third prediction gene set;

[0097] The prediction gene set acquisition unit is configured to compare the third prediction gene set with the fourth initial prediction gene set to obtain a fourth overlapping region gene set; and perform a fourth integration on the fourth overlapping region gene set to obtain a final prediction gene set.

[0098] In a preferred embodiment,

[0099] The electronic device also includes: a first labeling unit, used to record the initial comparison gene, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set as the gene sets to be compared; a second labeling unit, used to record the first initial prediction gene set, the first prediction gene set, the second prediction gene set and the third prediction gene set as the reference gene set; a third labeling unit, used to record the first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set as the overlapping region gene set;

[0100] The integration unit includes: a comparison module, which is used to compare the gene set to be compared with the reference gene set to obtain the gene set of the overlapping region;

[0101] A filtering module is used to filter the gene pairs with inconsistent directions between the to-be-compared gene set and the reference gene set in the overlapping region gene set, as well as the genes with UTR regions at both ends in the reference gene set, in the to-be-compared gene set to obtain a filtered gene set;

[0102] An extension module is used to extend each gene in the filtered gene set to obtain the extended region of each gene;

[0103] The extended region does not overlap with the positions of genes in the upstream region or the downstream region of the genes corresponding to the reference gene set;

[0104] The extended modules include:

[0105] The first extension submodule is used for extending the gene in the filtered gene set to the other end of the gene in the filtered gene set to obtain an extended region when one of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region;

[0106] The second extension submodule is used to extend to both ends of the gene to obtain the extended region when both ends of the gene in the filtered gene set do not contain UTR regions and both ends of the gene in the reference gene set corresponding to the gene in the filtered gene set have UTR regions:

[0107] A third extension submodule is used to extend upstream of the gene in the filtering gene set to obtain an extended region when the starting position of the gene in the reference gene set is smaller than the starting position of the corresponding gene in the filtering gene set;

[0108] A fourth extension submodule, for extending downstream of the gene in the filtering gene set to obtain an extension region when the end position of the gene in the reference gene set is smaller than the start position of the gene in the corresponding filtering gene set and the end position of the gene in the reference gene set is smaller than the end position of the gene in the corresponding filtering gene set;

[0109] Integrator module: Add the information of genes in the extended region to the reference gene set to obtain the final predicted gene set; the information of genes in the extended region includes the sequence position and size of the genes.

[0110] In a third typical embodiment of the present application, a processor is provided, which is used to run a program, and when the program is run, the above-mentioned gene prediction method for improving the completeness of gene UTR region prediction is executed.

[0111] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the present invention.

[0112] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus hardware devices such as detection devices. Based on such an understanding, the data processing part of the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application or some parts of the embodiments.

[0113] The present application can be used in many general or special computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0114] The method provided in this application can be executed in a terminal, a computer terminal or a similar computing device. Taking running on a terminal as an example, Figure 4 1 is a hardware structure block diagram of a terminal of a method for gene prediction for improving the completeness of gene UTR region prediction according to an embodiment of the present invention. Figure 4 As shown, the terminal may include one or more ( Figure 3Only one is shown in the figure) processor A1 (processor A1 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory B1 for storing data. Optionally, the terminal may also include a transmission device C1 and an input and output device D1 for communication functions. It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only and does not limit the structure of the above terminal. Figure 4 More or fewer components as shown, or with Figure 3 Different configurations shown.

[0115] The memory B1 can be used to store computer programs, for example, software programs and modules of application software, such as computer programs corresponding to methods such as read segment splicing, clustering, and consistency processing in the embodiments of the present invention. The processor A1 executes various functional applications and data processing by running the computer programs stored in the memory B1, that is, implementing the above method. The memory B1 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory B1 may further include a memory remotely arranged relative to the processor A1, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0116] The transmission device C1 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the terminal. In one example, the transmission device C1 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device C1 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0117] Obviously, those skilled in the art should understand that the above-mentioned partial modules or steps of the present application can be implemented in a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented with program codes executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module to implement. In this way, the present application is not limited to any specific combination of hardware and software. The beneficial effects of the present application will be further explained in detail below in conjunction with specific embodiments.

[0118] The beneficial effects of the present application will be further explained in detail below in conjunction with specific embodiments.

[0119] Example 1

[0120] The gene prediction method in the present application was applied to the gene prediction of the UTR region of the Papyrus chinensis and Papyrus mulberry genomes. The gene prediction methods of the two samples were consistent, so Papyrus chinensis and Papyrus mulberry were collectively referred to as samples.

[0121] The specific steps of gene prediction are as follows:

[0122] 1. Use hisat2 software to align the second-generation transcriptome sequencing data of roots, stems, leaves, flowers, and fruits at different stages of the samples to the genome of the samples;

[0123] 2. Use Stringtie software to assemble the results of the comparison in step 1 to obtain the gene annotation results of the roots, stems, leaves, flowers, and fruits of the samples at different stages.

[0124] 3. The above annotation results are merged to obtain a merged gene set, and the first transcriptome prediction is performed on the merged gene set using transdecoder software to obtain a first initial predicted gene set.

[0125] 4. Compare the first initial predicted gene set with the initial gene set to be compared, obtain the overlap situation (the first overlapping region gene set), and then integrate them to obtain the first predicted gene set.

[0126] 5. Use the third-generation transcriptome sequencing data of the samples to cluster and remove chimeras to obtain the third-generation full-length transcript data of the samples.

[0127] 6. After comparing the three-generation full-length transcript data of the sample with the genome of the sample using the PASA software program, the gene set obtained after the comparison was used for the second transcriptome prediction using the transdecoder software to obtain the second initial predicted gene set.

[0128] 7. Compare the second initial prediction gene set with the first prediction gene set, obtain the overlap between the two (second overlapping gene set), and then integrate them to obtain the second prediction gene set.

[0129] 8. Using liftoff software, according to the genome of homologous species and gene set annotation information, the genome of the sample is predicted to be homologous, and the third initial predicted gene set is obtained.

[0130] 9. Compare the third initial prediction gene set with the second prediction gene set, obtain the overlap between the two (third overlapping gene set), and then integrate them to obtain the third prediction gene set.

[0131] 10. Use minimap2 software to compare the data of the full-length transcripts downloaded from Refseq with the genome of the sample, and use transdecoder software to predict the gene structure of the gene set after the above comparison to obtain the fourth initial predicted gene set.

[0132] 11. Compare the fourth initial prediction gene set with the third prediction gene set, obtain the overlap between the two (the fourth overlapping gene set), and then integrate them to obtain the final prediction gene set.

[0133] The specific comparison and integration steps of the above steps 4, 7, 9, and 11 include:

[0134] The initial gene set to be compared, the second initial predicted gene set, the third initial predicted gene set and the fourth initial predicted gene set are recorded as the gene sets to be compared; the first initial predicted gene set, the first predicted gene set, the second predicted gene set and the third predicted gene set are recorded as the reference gene sets;

[0135] The first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set are recorded as overlapping region gene sets;

[0136] The integration includes: a) comparing the gene set to be compared with the reference gene set to obtain the overlapping region gene set; filtering the gene pairs in the overlapping region gene set that have inconsistent directions between the gene set to be compared and the reference gene set, as well as the genes with UTR regions at both ends of the reference gene set, in the gene set to be compared to obtain the filtered gene set;

[0137] b) Extend each gene in the filtered gene set to obtain the extended region of each gene; the extended region does not overlap with the position of the gene in the upstream region or the position of the gene in the downstream region of the gene corresponding to the reference gene set;

[0138] Wherein, when the 5' end of the corresponding gene in the reference gene set in the gene in the filtered gene set has a UTR region, the UTR region is extended toward the 3' end of the gene in the filtered gene set to obtain an extended region;

[0139] When the 3' end of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region, the UTR region is extended toward the 5' end of the gene in the filtered gene set to obtain an extended region.

[0140] The genes in the filtered gene set do not contain UTR regions at both ends, and the genes in the reference gene set corresponding to the genes in the filtered gene set have UTR regions at both ends, then:

[0141] The starting position of the gene in the reference gene set is smaller than the starting position of the corresponding gene in the filtering gene set, and the region is extended to the left end of the gene in the filtering gene set to obtain the extended region;

[0142] The end position of the gene in the reference gene set is smaller than the start position of the gene in the corresponding filtering gene set, and the end position of the gene in the reference gene set is smaller than the end position of the gene in the corresponding filtering gene set, and the extension is performed to the right end of the gene in the filtering gene set to obtain the extension region;

[0143] c) Compare the genes in the extended region with the gene set to be compared and the reference gene set respectively to obtain information about the genes in the extended region, and integrate the gene structure information of the extended region into the reference gene set. The information about the genes in the extended region includes the sequence position and size of the genes.

[0144] The merged gene structure information not only includes UTR regions but also may include exon / cds / intron regions, so that the gene set can be extended and improved.

[0145] Method for obtaining the initial gene set to be compared:

[0146] 1. Use RepeatMasker software and RepeatModeler software to predict the repetitive sequences of sample genomes.

[0147] According to the repeated sequence prediction results, the genomes of Paper mulberry and Broussonetia papyrifera were masked respectively to shield the repeated sequence regions and obtain the masked sample genome sequences;

[0148] 2. Use Stringtie software and Pasa software to predict the transcriptome of the sample genome sequence after masking to obtain the first sample prediction gene set;

[0149] Genewise software was used to predict homology of the sample genome sequence after masking to obtain the predicted gene set of the second sample;

[0150] August software was used to perform de novo prediction on the masked genome sequences of the papyrus chinensis and the papyrus pavers, respectively, to obtain the predicted gene set of the third sample;

[0151] The EVM software was used to integrate the first sample prediction gene set, the second sample prediction gene set, and the third sample prediction gene set to obtain the initial gene set.

[0152] Using the above method, the number of UTR regions of genes in the initial gene set to be compared (0), the first predicted gene set (1), the second predicted gene set (2), the third predicted gene set (3) and the predicted gene set (4) are shown in Table 1 (Paper mallow) and Table 2 (Paper mulberry).

[0153] Table 1

[0154] .

[0155] Table 2

[0156] .

[0157] From the above description, it can be seen that the above-mentioned embodiments of the present invention achieve the following technical effects: the gene prediction method of the present application significantly improves the accuracy and completeness of gene UTR region prediction by combining second-generation and third-generation sequencing technologies and utilizing information from homologous species and Refseq libraries, and can help technical personnel in this field to more accurately locate the non-coding regions of genes, which is of great value for understanding gene regulatory mechanisms, disease-related gene research, and genetic engineering applications.

[0158] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for gene prediction for improving the completeness of gene UTR region prediction, characterized in that, The method comprises: S1) performing a first transcriptome prediction, a second transcriptome prediction, a first homology prediction, and a third transcriptome prediction on the genome of the sample, respectively, to obtain a corresponding first initial prediction gene set, a second initial prediction gene set, a third initial prediction gene set, and a fourth initial prediction gene set; Wherein, the first initial predicted gene set is obtained by performing the first transcriptome prediction on the sample genome using the second-generation transcript data of the sample; The second initial predicted gene set is obtained by performing the second transcriptome prediction on the genome of the sample using the third-generation transcript data of the sample; The third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using a gene set in a homologous species genome database; The fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using a full-length transcript reference genome; S2) integrating the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set; The S2) includes: S21) comparing the first initial predicted gene set with the initial gene set to be compared to obtain a first overlapping region gene set; Performing a first integration on the first overlapping region gene set to obtain a first predicted gene set; The method for obtaining the initial gene set to be compared is as follows: performing sample gene prediction on the genome of the sample to obtain the initial gene set to be compared; S22) comparing the first predicted gene set with the second initial predicted gene set to obtain a second overlapping region gene set; performing a second integration on the second overlapping region gene set to obtain a second predicted gene set; S23) comparing the second predicted gene set with the third initial predicted gene set to obtain a third overlapping region gene set; Performing a third integration on the third overlapping region gene set to obtain a third predicted gene set; S24) comparing the third predicted gene set with the fourth initial predicted gene set to obtain a fourth overlapping region gene set; Performing a fourth integration on the fourth overlapping region gene set to obtain the final predicted gene set; S25) recording the initial gene set to be compared, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set as gene sets to be compared; Recording the first initial prediction gene set, the first prediction gene set, the second prediction gene set and the third prediction gene set as reference gene sets; The first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set and the fourth overlapping region gene set are recorded as overlapping region gene sets; The integration includes: a) comparing the gene set to be compared with the reference gene set to obtain the overlapping region gene set; Filtering the gene pairs in the overlapping region gene set that have inconsistent directions between the gene set to be compared and the reference gene set, and the genes with UTR regions at both ends of the reference gene set, in the gene set to be compared to obtain a filtered gene set; b) extending each gene in the filtered gene set to obtain an extended region of each gene; The extended region does not overlap with the position of the gene in the upstream region or the position of the gene in the downstream region of the gene corresponding to the reference gene set; in, One of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region, and the region is extended toward the other end of the gene in the filtered gene set to obtain the extended region; Both ends of the gene in the filtered gene set do not contain UTR regions, and both ends of the gene in the reference gene set corresponding to the gene in the filtered gene set have UTR regions, and the extended regions are obtained by extending to both ends of the gene: The extending comprises: when the starting position of a gene in the reference gene set is smaller than the starting position of a corresponding gene in the filtering gene set, extending upstream of the gene in the filtering gene set to obtain the extended region; When the end position of a gene in the reference gene set is smaller than the start position of a corresponding gene in the filtering gene set, and the end position of a gene in the reference gene set is smaller than the end position of a corresponding gene in the filtering gene set, extending downstream of the gene in the filtering gene set to obtain the extended region; c) adding information of genes in the extended region to the reference gene set, thereby obtaining the final predicted gene set; The information of the gene in the extended region includes the sequence position and size of the gene.

2. The method according to claim 1, characterized in that The first homology prediction software includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

3. The method according to claim 1, characterized in that The full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

4. The method according to claim 1, characterized in that: The software for performing the first transcriptome prediction, the software for performing the second transcriptome prediction, and the software for performing the third transcriptome prediction are each independently selected from PASA software and / or transdecoder software.

5. An electronic device for gene prediction for improving the completeness of gene UTR region prediction, characterized in that: The electronic device includes an initial prediction unit and an integration unit; The initial prediction unit is used to perform a first transcriptome prediction, a second transcriptome prediction, a first homology prediction and a third transcriptome prediction on the genome of the sample, respectively, to obtain a corresponding first initial prediction gene set, a second initial prediction gene set, a third initial prediction gene set and a fourth initial prediction gene set; Wherein, the first initial predicted gene set is obtained by performing the first transcriptome prediction on the sample genome using the second-generation transcript data of the sample; The second initial predicted gene set is obtained by performing the second transcriptome prediction on the genome of the sample using the third-generation transcript data of the sample; The third initial predicted gene set is obtained by performing a first homology prediction on the genome of the sample using a gene set in a homologous species genome database; The fourth initial predicted gene set is obtained by performing a third transcriptome prediction on the genome of the sample using a full-length transcript reference genome; The integration unit is used to integrate the first initial prediction gene set, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set to obtain a final prediction gene set; The integration unit includes a first prediction gene set acquisition unit, a second prediction gene set acquisition unit, a third prediction gene set acquisition unit and a prediction gene set acquisition unit; The first prediction gene set acquisition unit is used to compare the first initial prediction gene set with the initial gene set to be compared to obtain a first overlapping region gene set; and to perform a first integration on the first overlapping region gene set to obtain the first prediction gene set; The initial gene set to be compared is obtained by an initial gene set to be compared acquisition unit, wherein the initial gene set to be compared acquisition unit is used to perform sample gene prediction on the genome of the sample to obtain the initial gene set to be compared; The second prediction gene set acquisition unit is used to compare the first prediction gene set with the second initial prediction gene set to obtain a second overlapping region gene set; perform a second integration on the second overlapping region gene set to obtain the second prediction gene set; The third prediction gene set acquisition unit is used to compare the second prediction gene set with the third initial prediction gene set to obtain a third overlapping region gene set; perform a third integration on the third overlapping region gene set to obtain the third prediction gene set; The prediction gene set acquisition unit is configured to compare the third prediction gene set with the fourth initial prediction gene set to obtain a fourth overlapping region gene set; and perform a fourth integration on the fourth overlapping region gene set to obtain the final prediction gene set; The electronic device further comprises: A first labeling unit, used for marking the initial comparison gene, the second initial prediction gene set, the third initial prediction gene set and the fourth initial prediction gene set as gene sets to be compared; a second labeling unit, configured to record the first initial prediction gene set, the first prediction gene set, the second prediction gene set, and the third prediction gene set as reference gene sets; a third labeling unit, configured to record the first overlapping region gene set, the second overlapping region gene set, the third overlapping region gene set, and the fourth overlapping region gene set as overlapping region gene sets; The integration unit comprises: A comparison module, used for comparing the gene set to be compared with the reference gene set to obtain the overlapping region gene set; A filtering module, for filtering the gene pairs in the overlapping region gene set that have inconsistent directions between the gene set to be compared and the reference gene set, and the genes with UTR regions at both ends of the reference gene set, in the gene set to be compared, to obtain a filtered gene set; An extension module, used for extending each gene in the filtered gene set to obtain an extended region of each gene; The extended region does not overlap with the position of the gene in the upstream region or the position of the gene in the downstream region of the gene corresponding to the reference gene set; Wherein, the extension module comprises: A first extension submodule is used for extending the gene in the filtered gene set toward the other end of the gene in the filtered gene set to obtain the extended region when one of the two ends of the gene in the reference gene set corresponding to the gene in the filtered gene set has a UTR region; The second extension submodule is used for extending to both ends of the gene to obtain the extended region when both ends of the gene in the filtered gene set do not contain UTR regions and both ends of the gene in the reference gene set corresponding to the gene in the filtered gene set have UTR regions: A third extension submodule is used for, when the starting position of a gene in the reference gene set is smaller than the starting position of a corresponding gene in the filtering gene set, extending upstream of the gene in the filtering gene set to obtain the extended region; a fourth extension submodule, configured to extend the gene in the filtering gene set downstream to obtain the extended region when the end position of the gene in the reference gene set is smaller than the start position of the corresponding gene in the filtering gene set and the end position of the gene in the reference gene set is smaller than the end position of the corresponding gene in the filtering gene set; An integration submodule, used to integrate the information of the genes in the extended region into the reference gene set, so as to obtain the final predicted gene set; The information of the gene in the extended region includes the sequence position and size of the gene.

6. The electronic device according to claim 5, characterized in that: The first homology prediction software includes one or more of liftoff software, genewise software, Gemoma software or miniprot software.

7. The electronic device according to claim 5, characterized in that: The full-length transcript reference genome is derived from the Refseq database, the EMBL database, the NGDC database or the GENCODE database.

8. The electronic device according to claim 5, characterized in that: The software for performing the first transcriptome prediction, the software for performing the second transcriptome prediction, and the software for performing the third transcriptome prediction are each independently selected from PASA software and / or transdecoder software.

9. A computer-readable storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the method for gene prediction for improving the completeness of gene UTR region prediction according to any one of claims 1 to 4 is controlled.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the method for gene prediction for improving the completeness of gene UTR region prediction according to any one of claims 1 to 4.

Images